Covid Report for Family

Background

Notes and format last updated May 7, 2020

Starting on the May 7th update, the NY Times began including probable covid cases/deaths along with confirmed. This mostly affects death counts – for certain geographies that include probable COVID deaths in addition to confirmed, these are now added to the totals. For the time being, they were all added to the May 6th totals, causing a big spike at the U.S. level. Over time, NY Times will revise their historical counts and distribute these added deaths when they actually occurred, so the spike should fade.

• This page shows COVID data broken down into regions that affect our family. It pulls from publicly available raw data from the NY Times, which is reported daily. It’s on a one-day lag, so when you see this report, the most recent day included will be that reported the day prior.
• The table of contents above summarizes the major sections and can be used to jump to those sections.
• Each major section is then divided into tabs to allow you to more easily shuffle between geographies and metrics.
  • If you open on your phone, I don’t believe the tabs will work, so you’ll just see the page as one long, continuous document.
• More detailed descriptions and context can be found in each section.

Growth rates

Heat maps

• The two heat maps below compare how quickly total cases or deaths have grown at various times in our respective geopgraphies.
• The first plot compares growth rate for total cases; the second, growth rate for total deaths.
• The metric used is doubling time, by which I mean how quickly total cases or deaths are doubling.
• The plots track that doubling time at each date for our geographies. Darker colors reflect shorter doubling times, and thus periods of faster growth.
  • You can use the plots to track each geography over time and to compare the geographies to one another.
  • You can also compare the cases and death charts, to see how faster periods of death growth follow faster periods of case growth.

Case growth rates

• This section charts the growth rate of both total and new cases for each of our respective geographies. Each geography has its own chart, and then that chart will have a trendline for total cases and new cases.
  • There are only plots for the U.S. and states because the numbers for the counties are too small to generate worthwhile trendlines in this section.
• Note that we’re charting growth rate and not a count of cases, so don’t think of these as the standard “curve” that we hear about in the news and that we want to flatten. Instead, these growth rate charts help track more precisely what we can only estimate when we see those other curves. For these growth rate charts, if the line is above zero, the metric we are tracking (total or new cases) is continuing to grow. If the growth rate line is going up, it’s growing more quickly each day; if it’s going down but still above zero, it’s growing less quickly (but still growing). Only when the growth rate lines go below zero has the metric stopped growing.
• Each of these two lines uses rolling windows to calculate a growth rate for that particular metric. I do the calculation differently for each to smooth out some of the large day-to-day discrepancies in new case reporting at the state level.
  • For total cases, the trendlines are a rolling 3-day average of daily growth rates in total cases. We want to see these decline (and almost all are), but they can’t go below zero. This is because we’re tracking growth rate and a growth rate line below zero would mean total cases have gone down, which can’t happen. They can only grow less quickly, which means we want to see the total case line get as close to zero as possible.
  • For new cases, the trendlines show a rolling 3-day average of daily growth rate in the rolling 7-day average of new cases. Including two rolling periods in this average helps smooth out crazy spikes at the state level that result from large day-to-day changes. Unlike the lines for total cases, we want to watch for the lines for new cases to get consistently below zero and stay there. That means that we are consistently seeing fewer new cases on a daily basis.

U.S.

Our states

Death growth rates

• This section charts the growth rate of both total and new deaths for each of our respective geographies. Each geography has its own chart, and then that chart will have a trendline for total deaths and new deaths.
  • There are only plots for the U.S. and states because the numbers for the counties are too small to generate worthwhile trendlines in this section.
• Note that we’re charting growth rate and not a count of deaths, so don’t think of these as the standard “curve” that we hear about in the news and that we want to flatten. Instead, these growth rate charts help track more precisely what we can only estimate when we see those other curves. For these growth rate charts, if the line is above zero, the metric we are tracking (total or new deaths) is continuing to grow. If the growth rate line is going up, it’s growing more quickly each day; if it’s going down but still above zero, it’s growing less quickly (but still growing). Only when the growth rate lines go below zero has the metric stopped growing.
• Each of these two lines uses rolling windows to calculate a growth rate for that particular metric. I do the calculation differently for each to smooth out some of the large day-to-day discrepancies in new death reporting at the state level.
  • For total deaths, the trendlines are a rolling 3-day average of daily growth rates in total deaths. We want to see these decline (and almost all are), but they can’t go below zero. This is because we’re tracking growth rate and a growth rate line below zero would mean total deaths have gone down, which can’t happen. They can only grow less quickly, which means we want to see the total death line get as close to zero as possible.
  • For new deaths, the trendlines show a rolling 3-day average of daily growth rate in the rolling 7-day average of new deaths. Including two rolling periods in this average helps smooth out crazy spikes at the state level that result from large day-to-day changes. Unlike the lines for total deaths, we want to watch for the lines for new deaths to get consistently below zero and stay there. That means that we are consistently seeing fewer new deaths on a daily basis.

U.S.

Our states

By population rankings

This section tracks metrics for states and counties normalized for population (number of cases or deaths per million residents), and then compares these figures both for our geographies and the country overall.

States

• This section shows tables ranking all 50 states for per populations rates of total cases, new cases, total deaths, and new deaths.
• For each metric, in addition to the tables, the trends for the top states are plotted over time.
  • We only plot the top ten states for each metric so that the plots aren’t too crowded. But you can view the full 50-state rankings in the tables.

Total confirmed cases

Table of total confirmed cases per million residents (all 50 states)

Ranking	State	Cases Per Million
1	North Dakota	108,897
2	South Dakota	97,202
3	Iowa	77,554
4	Wisconsin	75,753
5	Nebraska	72,287
6	Utah	67,189
7	Montana	63,507
8	Wyoming	62,749
9	Illinois	62,257
10	Minnesota	62,224
11	Idaho	61,862
12	Kansas	58,914
13	Rhode Island	58,655
14	Tennessee	57,311
15	Indiana	57,097
16	Arkansas	56,638
17	Mississippi	55,417
18	Missouri	55,060
19	Alabama	55,041
20	Nevada	54,788
21	Oklahoma	54,710
22	Louisiana	54,018
23	New Mexico	51,548
24	Alaska	50,368
25	Arizona	50,046
26	Florida	49,263
27	Texas	46,358
28	Kentucky	45,834
29	Colorado	45,575
30	Georgia	45,501
31	South Carolina	45,079
32	Michigan	42,668
33	New Jersey	41,639
34	Delaware	40,987
35	Ohio	40,638
36	North Carolina	37,663
37	Massachusetts	37,264
38	New York	36,528
39	Connecticut	35,821
40	Maryland	35,620
41	California	34,721
42	Pennsylvania	33,106
43	District of Columbia	32,782
44	West Virginia	30,687
45	Virginia	29,881
46	Puerto Rico	28,270
47	Washington	24,360
48	Oregon	20,033
49	New Hampshire	18,217
50	Hawaii	13,123
51	Maine	9,929
52	Vermont	8,037

New confirmed cases

Table of new cases per million residents: rolling 3-day average (all 50 states)

Ranking	State	New Cases Per Million
1	Alaska	1,125
2	Indiana	1,099
3	Oklahoma	1,047
4	Minnesota	1,018
5	South Dakota	995
6	Utah	951
7	Nevada	913
8	Nebraska	882
9	Delaware	840
10	Wisconsin	832
11	New Mexico	829
12	Montana	821
13	North Dakota	820
14	Pennsylvania	812
15	Ohio	803
16	Colorado	791
17	Massachusetts	770
18	Kentucky	740
19	Arizona	739
20	West Virginia	738
21	Arkansas	730
22	Illinois	728
23	Wyoming	713
24	Kansas	689
25	Idaho	684
26	Missouri	673
27	Mississippi	660
28	California	658
29	Alabama	647
30	New Jersey	638
31	Iowa	630
32	South Carolina	585
33	Tennessee	571
34	New York	543
35	Maryland	531
36	North Carolina	529
37	Michigan	525
38	Louisiana	505
39	District of Columbia	459
40	New Hampshire	452
41	Florida	450
42	Rhode Island	445
43	Georgia	412
44	Virginia	412
45	Oregon	411
46	Texas	386
47	Puerto Rico	267
48	Washington	235
49	Vermont	199
50	Maine	196
51	Connecticut	143
52	Hawaii	76

Total deaths

Table of total deaths per million residents (all 50 states)

Ranking	State	Deaths Per Million
1	New Jersey	1,950
2	New York	1,776
3	Massachusetts	1,596
4	Connecticut	1,443
5	Louisiana	1,416
6	North Dakota	1,337
7	Rhode Island	1,333
8	Mississippi	1,330
9	South Dakota	1,254
10	Illinois	1,114
11	Michigan	1,032
12	District of Columbia	987
13	Arizona	954
14	Indiana	927
15	Georgia	896
16	Florida	892
17	Pennsylvania	889
18	South Carolina	886
19	Arkansas	881
20	Iowa	861
21	New Mexico	834
22	Delaware	814
23	Maryland	801
24	Texas	799
25	Alabama	793
26	Nevada	752
27	Missouri	720
28	Minnesota	716
29	Tennessee	715
30	Montana	688
31	Wisconsin	679
32	Nebraska	627
33	Kansas	613
34	Colorado	596
35	Ohio	595
36	Idaho	581
37	North Carolina	531
38	California	504
39	Kentucky	503
40	Virginia	492
41	Oklahoma	479
42	West Virginia	467
43	Wyoming	444
44	New Hampshire	414
45	Washington	402
46	Puerto Rico	372
47	Utah	292
48	Oregon	244
49	Alaska	185
50	Hawaii	183
51	Maine	168
52	Vermont	126

New deaths

Table of new deaths per million residents: rolling 3-day average (all 50 states)

Ranking	State	New Deaths Per Million
1	South Dakota	29
2	North Dakota	15
3	Illinois	12
4	Iowa	12
5	Kansas	12
6	New Mexico	12
7	Arkansas	11
8	Minnesota	11
9	Indiana	10
10	Wisconsin	10
11	Michigan	9
12	Mississippi	9
13	Pennsylvania	9
14	West Virginia	9
15	Alabama	7
16	Missouri	7
17	Tennessee	7
18	Massachusetts	6
19	Nebraska	6
20	Nevada	6
21	South Carolina	6
22	Texas	6
23	Alaska	5
24	Arizona	5
25	Colorado	5
26	Ohio	5
27	Oklahoma	5
28	Delaware	4
29	Florida	4
30	Georgia	4
31	Kentucky	4
32	Louisiana	4
33	Maryland	4
34	Montana	4
35	New Hampshire	4
36	New Jersey	4
37	Oregon	4
38	Rhode Island	4
39	Connecticut	3
40	Hawaii	3
41	Idaho	3
42	New York	3
43	North Carolina	3
44	Puerto Rico	3
45	California	2
46	District of Columbia	2
47	Utah	2
48	Vermont	2
49	Virginia	2
50	Maine	1
51	Washington	1
52	Wyoming	0

Counties

• This section focuses on the county level. It shows tables with our counties ranked by percentile of U.S. counties for per population rates of total cases and total deaths.
  • Each table also shows the top five counties in the country in addition to our counties, for added perspecive.
• In addition to the tables, our counties’ percentile for both total cases and total deaths are plotted over time.

Confirmed cases

Table showing total cases per million and percentile for all US counties. Includes our counties and the top 5 in the US for perspective.

County	State	Cases Per Million	Raw Ranking	Percentile
Crowley	Colorado	221,086	1	99
Norton	Kansas	209,289	2	99
Bon Homme	South Dakota	203,014	3	99
Buffalo	South Dakota	199,286	4	99
Lincoln	Arkansas	197,251	5	99
Davidson	Tennessee	72,151	504	83
Richland	South Carolina	51,729	1321	57
York	South Carolina	38,579	2048	34
Orange	California	28,080	2539	19
Pierce	Washington	21,826	2762	12

Our county percentiles over time

Deaths

Table showing total deaths per million and percentile for all US counties. Includes our counties and the top 5 in the US for perspective.

County	State	Deaths Per Million	Raw Ranking	Percentile
Gove	Kansas	7,587	1	99
Jerauld	South Dakota	6,458	2	99
Dickey	North Dakota	5,952	3	99
Hancock	Georgia	5,558	4	99
Gregory	South Dakota	5,496	5	99
Richland	South Carolina	736	1418	54
Davidson	Tennessee	631	1651	47
York	South Carolina	527	1879	40
Orange	California	514	1911	39
Pierce	Washington	366	2277	27

Our county percentiles over time

Raw counts

• This section shows the raw counts over time for the U.S., our states, and our counties for four categories:
  • Total confirmed cases: This is the cumulative total of cases reported.
  • New confirmed cases: This is the number of new cases reported on each respective date.
    • For the U.S. and states, a trendline shows the rolling 7-day average of new cases, to help better visualize the peak.
  • Total deaths: This is the cumulative total of deaths reported.
  • New deaths: This is the number of new deaths reported on each respective date.
    • For the U.S. and states, a trendline shows the rolling 7-day average of new deaths, to help better visualize the peak.

Total confirmed cases

U.S.

Our states

Our counties

New confirmed cases

U.S.

Our states

Our counties

Total deaths

U.S.

Our states

Our counties

New deaths

U.S.

Our states

Our counties

Stay-at-home comparisons

• This section is a work in progress. It’s exploring what effect stay-at-home orders have on spread. It’s inherentely messy because some states have localized orders but not orders for their full state and because states implemented the orders at different times.
  • The first plot shows a rolling 3-day average of new cases per million residents, averaged across 2 groups of states – those with a state-wide stay-at-home order in effect and those without. The size of the dots reflect the number of states in each group for each given day (so the dots are getting bigger over time for the group with orders, as more states implement those orders).
    • So far, this view doesn’t show if these orders are “working” or not. Instead, the biggest takeaway from the current plot is that the states that implemented orders are reacting to having worse spread, which is why the case rate is so much higher at the start for them. As a state decides it has a problem and implements an order, it joins the grouping of states with orders and drives the rate up for that group.
    • Over time (if the policies work), what we’d expect is that the line for the group with orders flattens or declines in relation to the no-order group. It also may become moot if all states issue orders.
    • As states start rescending their orders, I’ll put them back into the no-order group, so the no-order group should beging to expand again.
  • The second plot tracks progress of states by the number of days since orders have been in effect. It includes only those states with orders and shows their average new case growth rate (a rolling 3-day average of daily growth rates) based on how long it’s been since the order went into effect. Again, the size of the dots show how many states are included in each point – i.e., as you go further right, the dots are smaller because less states are that far past their order.
    • I’m working on how to add a baseline to this chart to compare to states without orders, but that’s challenging because there’s no way to fully align timing.